Study Assesses Outcomes of Balloon Atrial Septostomy on Neonates With Transposition of the Great Arteries

Transposition of the great arteries (TGA) is a common congenital heart defect requiring surgical correction in the first week of life. Several decades’ experience with the arterial switch procedure has dropped surgical mortality to 2.8%, a figure that continues to improve. However, presurgical mortality remains higher.

About 12% of TGA patients undergo urgent balloon atrial septostomy (BAS) to improve oxygenation until definitive surgical treatment can be performed. BAS is associated with stroke, but whether the procedure actually causes stroke is a matter of debate. Pediatric critical care medicine specialists at Cleveland Clinic Children’s sought to clarify this issue by comparing outcomes of BAS and no-BAS in patients with TGA. The results were not exactly as expected.

“We did not find a significant statistical difference in mortality between BAS and no-BAS patients. However, we did identify factors associated with increased in-hospital mortality risk in specific subgroups of BAS and no-BAS patients,” says Mohammed Hamzah, MD, first author of the study published in Pediatric and Critical Care Medicine. “These findings suggest that a change in delivery protocol for babies diagnosed with TGA in utero could reduce preoperative mortality risk.”

How the study was done

The researchers reviewed data on 17,392 neonates in two national databases, including the Kids’ Inpatient Database, the largest publicly available pediatric database in the United States. Neonates age 28 days or less with the diagnosis of TGA were identified and stratified into those who underwent BAS (27.7%) and those who had not.

Primary outcomes were death before hospital discharge, length of stay (LOS) and stroke during hospitalization. Secondary outcomes were extracorporeal membrane oxygenation (ECMO) support during hospitalization and necrotizing enterocolitis (NEC).

Although mortality rates were similar in BAS and no-BAS groups, BAS patients had significantly longer LOS and significantly higher rates of ECMO and stroke.

Where BAS and no-BAS diverged

A separate analysis of patients who died during hospitalization (6.6%) revealed statistically significant differences between the BAS and no-BAS groups in three areas: transferred-in status, ECMO utilization and stroke.

A higher percentage of transferred-in patients and those requiring ECMO were in the no-BAS group. More strokes were seen in the BAS group. All three variables were found to increase mortality risk.

Although a retrospective study could not explain these findings, Dr. Hamzah hypothesized that the transferred-in patients were too sick to undergo BAS. “This hypothesis is somewhat supported by the association between ECMO use and mortality risk in the no-BAS group,” he says.

Lowering “hidden mortality”

Dr. Hamzah calculated that the reason for about 3% of presurgical mortality is unknown. He suggests the reason for this “hidden mortality” is the decline in patient status during transfer to the center where TGA surgery would be performed.

“When we looked at the babies born elsewhere and transferred in, their mortality rate was higher than those whose diagnosis was known before birth, and the birth occurred at the hospital where the TGA was treated,” he says.

“It’s important for obstetricians to recognize that the decision whether to perform BAS is usually made within 24 hours of birth. When TGA has been diagnosed prenatally, the mother should be transferred to the hospital where the baby will be treated while it is in utero. This will allow BAS to be done in a safe, stable environment, which should decrease the presurgical mortalilty risk,” he says.